Mass spectrometry (MS) is a well-known technique of obtaining a molecular weight and structural information about chemical compounds. Using mass spectrometry techniques, molecules may be weighed by ionizing the molecules and measuring the response of their trajectories in a vacuum to electric and magnetic fields. Ions are weighed according to their mass-to-charge (m/z) values.
Atmospheric pressure ion sources (API) have become increasingly important as a means for generating ions used in mass spectrometers. Some common atmospheric pressure ion sources include Electrospray or nebulization assisted Electrospray (ES), Atmospheric Pressure Chemical Ionization (APCI), Atmospheric Photo Ionization (APPI), and Matrix Assisted Laser Desorption Ionization (MALDI). These ion sources produce charged particles, such as protonated molecular ions or adduct, from analyte species in solution or solid form, in a region which is approximately at atmospheric pressure.
API sources are advantageous because they provide a gentle means for charging molecules without inducing fragmentation. They also provide ease of use because samples can be introduced at atmosphere.
Mass spectrometers, however, generally operate in a vacuum maintained between 10−4 to 10−10 Torr depending on the mass analyzer type. Thus once created, the charged particles must be transported into vacuum for mass analysis. Typically, a portion of the ions created in the API sources are entrained in a bath gas API source chamber and swept into vacuum along with a carrier gas through an orifice into vacuum. Doing this efficiently presents numerous challenges.
Disadvantageously, API sources produce high chemical background and relatively low sensitivity. This results in a poor signal-to-noise ratio. This is believed to be caused by sampling of impurites attached to analyte ions (for example, cluster molecules, atoms or ions, or other undesired adducts), caused by incomplete desolvation during the API process. Many solvated droplets enter into the mass spectrometer and consequently produce a large level of chemical noise across the entire mass range. Additionally incompletely vaporized droplets linger near the sampling orifice.
These problems can be most severe for high flow rates. Efficient Electrospray Ionization (ESI) at high liquid flow rates requires sufficient energy transfer for desolvation and a method to deter large clusters from entering the vacuum chamber while enhancing the ion capture. High flow rate analyses are important to industries that have large throughput requirements (such as drug development today, and in the future, protein analysis). For most modern applications of ESI and APCI, liquid samples are passed through the source at high flow rates.
Another problem with electrospray concerns the condensation of the expanding jet and clustering of the ions. Various instrument manufactures use a conventional molecular beam interface to couple an ion source to the low pressure vacuum region. Conventionally, a molecular free jet is formed as gas expands from atmosphere into an evacuated region. The ion flux is proportional to the neutral density in a free jet, which depends on the shape and size of the orifice through which the gas expands, as well as the pressure of the evacuated region. In conventional ion sources, a skimmer samples the free jet, and the ions are detected downstream. This approach has several negative side effects, including: a) restricting the time for ion desolvation, b) enhancing ion solvation, c) restricting the gas flow through the orifice due to pumping requirements and the spatial requirements of sampling a free jet expansion.
To reduce the problem of incomplete desolvation, heated gases are commonly employed to vaporize with a flow direction opposite, or counter, to sprayed droplets in order to desolvate ions at atmospheric pressure. Since the heated gases remove some of the solvent vapor from the stream of gas before being drawn into the vacuum chamber, this technique may partially assist to increase the concentration of ions of interest entering the vacuum chamber.
While the counter flow of gas results in some improvement in sensitivity for low liquid flow rates, it is insufficient for high liquid flow rates, for example 10 microliters per minute or more, where substantially more energy transfer is required than the counter flow of gas can provide. Also, even for low liquid flow rates, it substantially increases the complexity of the interface between the electrospray and the mass spectrometer. In order that the solvent vapor from the evaporating droplets be efficiently removed by the counter flowing gas, both the temperature and the flow rate of the gas must be carefully controlled. High gas flow rates may prevent some ions with low mobility from entering the analyzer, while low gas flow rates or reduced gas temperature may not sufficiently desolvate the ions. The counter flowing gas flow rate and temperature are typically optimized for each analyte and solvent. Accordingly, much trial and error time is necessary to determine the optimum gas flow rate and temperature for each particular analyte utilizing a particular electrospray device and a particular mass spectrometer. As a result only a small fraction of the produced ions are focused by the lenses and transmitted to the mass analyzer for detection. Accordingly, this reduced transfer of ions to the mass analyzer produced by electrospray substantially limits the sensitivity and the signal-to-noise ratio of the electrospray/mass spectrometer technique.
Alternatively, an additional heated desolvation chamber located downstream of the first nozzle of a conventional molecular beam interface may be used. The electrosprayed droplets first expand in a supersonic expansion and then are passed into a second heated chamber pumped by a separate pumping system, which is maintained at a pressure preferably less than 1 Torr. This beam is then passed on-axis into a mass spectrometer. This design suffers from incomplete desolvation due to low residence time in the chamber, and compromises sensitivity due to scattering losses. Also the molecular beam is sampled on-axis with respect to the gas in the heated chamber, and therefore still permits incompletely de-solvated ions to enter the mass spectrometer. This design yields increased complexity and cost of an additional pumping stage following the initial expansion.
It is therefore desirable to provide an improved mass spectrometer interface for atmospheric pressure ionization sources.